Previous Page Print
Font size：
Small
Medium
Large

Amazing Smart Aquaculture : successfully reversing traditional farming

2024-04-10

The aquaculture industry has been developed in Taiwan for more than 300 years, and Taiwan's aquaculture technology has earned the reputation of "Kingdom of Aquaculture". However, in recent years, due to the abnormal climate and environment, limited land and water resources, drug residues affecting the food safety of aquatic products, market competition, increased cost of aquaculture, and dysfunctions in production and marketing, the development of the traditional aquaculture industry has stagnated, and it is also facing Labor shortage and aging problems.

In response to the problems encountered by traditional aquaculture, we have developed intelligent farming technologies, and have successively developed various smart technologies such as "Mobile Fish Counting and Length Measurement System", "Automated Total Bacteria Detection System", "Fish Farm Disaster Overhead Image Processing Technology", "Aquaculture Intelligent Expert System", and "Anemone Fish Culture Environment Monitoring and Early Warning Technology", with the goal of conducting The goal is to cultivate and raise healthy and high-quality aquaculture species, to assist the traditional aquaculture industry to develop towards production automation and intelligent management technology, to improve production efficiency, to further achieve productivity enhancement, regulation and planned production effects, and to promote intelligent aquaculture technology to the public and become the mainstream of Taiwan's aquaculture fishery in the future.

The research team uses the Internet of Things, video technology, information and communication technology, and intelligent interconnection concepts to develop environmental control systems, water quality microclimate network sensing and control technology, and intelligent management technology, which can effectively reduce human resources to achieve labor-saving benefits and reduce the risk of aquaculture, and ensure the effective transmission of relevant intelligent aquaculture experience.

After the introduction of technology, the production capacity of fish farming species and stages can be increased by 150%~200%, labor can be saved by 50%, water can be saved by more than 30%, feed costs can be reduced by at least 15%, and the deterioration of water quality in large fish farms can be forewarned 3 days earlier.

Title

Introduction

Design of Fish and Shrimp Counting and Size Measuring System

 Using the developed core image processing algorithm technology, combined with the complex mechanism and human-machine interface control software, it is further developed into an automated system equipment that can be effectively put into the process of aquaculture to process the number counting and body length measurement of fish or shrimp in the aquaculture pond to achieve the following objectives: (1) reduce the overall labor demand (2) more accurate and fast control of the growth status ( numbers and body length) of the aquaculture organisms (fish or shrimp).

Fast detection device for Vibrio maritimus digitization

 It can be mainly used for water quality inspection and abnormality warning in the actual farming environment, reducing the inspection time and labor cost of laboratory personnel in the farm, and obtaining the pathogen concentration data of each pool mechanically.

Automated total bacteriological count detection system

 The system uses micron-scale flow channels to precisely control the direction and volume of the fluid. The microfluidic chip is designed and manufactured based on the development blueprint of Vibrio maritimus MV-kit, and the arduino controller automatically controls the sample to be tested, buffer, kit and cleaning solution into the microfluidic chip, and digitizes the measurement value to record in the cloud, and notifies the app of the Vibrio concentration in the aquaculture water in time.

Versatile automated optical digital pathogen detection technology

 The system is designed to be fast, convenient, and lightweight, without the need for complicated optical instruments, excessive time and labor, and a certain degree of accuracy. Through the multi-band light source optical detection and test results can be recorded functional module design, combined with digital pathogen identification technology, to produce a multi-species universal pathogen detection automation system, providing real-time pathogen monitoring.

Aquaculture sensing networking decision feedback control methodology

 The remote controller and human-machine interface are designed and developed to effectively manage the aquaculture pond through real-time monitoring of water quality sensing (main water quality parameters, oxygenation and other equipment automation), and the signal analysis and processing core technology to summarize the aquaculture variation factors, teach back the automatic control database source, and provide statistics and analysis of aquaculture parameters.

Core technology of aquaculture intelligent expert system

 The system adopts intelligent robot technology and integrates natural language processing technology to perform word breaking, association analysis and guided keyword functions, and provides users with easy access to relevant information on both web and cell phones. By entering a question or keyword into the system, the system will answer the question with a reference answer and provide a series of questions related to the question, which can guide users to explore a topic in depth.

Fish Farm Damage High Altitude Image Processing Technology

 The deep learning model used is the pix2pix GAN training model. The model is trained by mutual generation of TGOS aerial orthoimagery and the corresponding e-map. The feasibility assessment of water body area identification is performed. (A total of 1456 training images were generated, and after rotation and mirroring, a total of 11,648 images were generated. The model trained by this technique can effectively mark the dike and water area in fish farms, and then achieve comparative marking of the flooded area.

3D underwater fish measurement system for aquaculture

 This technology is specially developed for high value ornamental aquarium species of institutionalized fish fry, and introduces 3D image information to reduce the error caused by various limitations of 2D images in measuring fish length. The 3D image acquisition module is built and adjusted to obtain images of fish bodies in the water, and the depth information of fish bodies is processed and estimated by combining the color segmentation and contour features with image analysis methods.

Environmental monitoring and advance warning technology for fish breeding

 This technology is mainly developed as a system software, which can be applied to the statistics and analysis of aquaculture parameters in aquaculture farms, to reduce the production water power and labor operation cost, to improve the stability of aquaculture environment water quality, to provide operators with healthy management of aquaculture fish species and to reduce aquaculture risks. This technology provides the anemone fish breeding environment monitoring and advance warning back to the automatic control database source through various environmental parameters statistical analysis, including (water temperature, pH, dissolved oxygen, salinity, ORP and environmental microclimate conditions, etc.) parameters, and signal analysis and processing of core technology variables, and provides aquaculture parameters statistics and analysis to reduce aquaculture production water power and labor operation cost. It also improves the stability of water quality in the aquaculture environment.

The intelligent fish feeding and fish dynamic identification technology for cage culture

 The technology is developed to monitor the fish growth and body size of adult fish in marine cage culture, analyze the most suitable feed placement process. The study using image recognition technology utilizes the ability to analyze and identify biological activity images to assist in determining feed switching and unit feed volume, thereby providing a referenced for precise feed throw management, control the amount of feed throw, improve cage culture production efficiency, and reduce excessive feed cost. At the same time, to reduce the cage culture production costs and waste offered, but also to reduce the pollution arising out of the sea environment.
TOP